Mucosal transmission of HIV involves a strong bottleneck effect and most often, a single founder virus is transmitted. The relative contributions of stochastic (random) selection vs. active selection by specific viral and host characteristics to this bottleneck are unclear. Mother-to-child HIV-1 transmission (MTCT) through breastfeeding, in which transmission pairs and timing of infant infection are readily identified, provides an instructive model to address these important and unresolved questions. We will use serial samples from a large, well characterized cohort of HIV-1 infected Zimbabwean women who transmitted HIV-1 through breastmilk. This cohort includes women with chronic HIV-1 infection (CI; N=35), and women who acquired primary HIV-1 infection post-partum (acute infection, AI; N=13). We combine phylogenetics (with frequency analyses based on deep sequencing data) with functional assays (CD4 and co-receptor use, entry kinetics, neutralization sensitivity), to quantitatively and comprehensively analyze the relationship of founder viruses to maternal blood or breastmilk variants, quantify the relative contributions of stochastic vs. active selectio, identify selection pressures on envelope (env), differentiate when selection pressures may operate during the transmission bottleneck, and determine whether founder env variants are better adapted than other non- transmitted variants for postnatal MTCT. Delineation of the biologic properties of transmitted variants along with mapping the genetic bases of these biologic properties, should improve understanding of HIV-1 entry and the mechanisms of action of HIV-1 entry inhibitors. These studies will also improve our understanding of the in vivo selective pressures exerted by autologous neutralizing antibodies and innate factors in two distinct and relevant anatomic compartments (blood, where levels of antibodies are high, and breastmilk, where they are much lower) and the potential role of neutralization sensitivity or escape in transmission. Finally, these studies will specifically reveal whether increasing the neutralizing activity of blood or breastmilk (e.g., through passive or active immunization) or targeting other Env functional properties (tropism, infectivity) hold promise to block primary infection of women and children.